APL-TDA acoustic loudspeaker analyzing software - diyAudio
Go Back   Home > Forums > Design & Build > Software Tools
Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Software Tools SPICE, PCB CAD, speaker design and measurement software, calculators

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 20th January 2016, 08:21 PM   #1
Juhazi is offline Juhazi  Finland
diyAudio Member
 
Juhazi's Avatar
 
Join Date: Mar 2007
Location: Jyväskylä, Finland
Default APL-TDA acoustic loudspeaker analyzing software

The APL-TDA software is a new acoustic analyzer software, which is aimed for professional sound technicians as a fast and focused tool for setting up loudspeaker systems.

it focuses in Time Delay Analysis (TDA) and uses bandpassed impulse response instead of usual FFT that eg. REW uses. Demo version is free and one must download a MATLAB compiler runtime package too.

wesayso brought it to this discussion about line arrays Corner Floor-to-Ceiling Line Array Using Vifa TC9

Now we are trying to learn what extra value it gives to diy enthusiasts. The demo version has mostly fixed parameters, bu I will try to compare it to RoomEQWizard.

Here as a couple of graphs from the website
TDA 3D presentation (decay)
Click the image to open in full size.
DFR (group delay per section ??)
Click the image to open in full size.
__________________
AINOgradient speaker project
  Reply With Quote
Old 20th January 2016, 08:36 PM   #2
Juhazi is offline Juhazi  Finland
diyAudio Member
 
Juhazi's Avatar
 
Join Date: Mar 2007
Location: Jyväskylä, Finland
I run quick tests of my desktop speaker (Creative Gigaworks T20, left) at roughly 1 meter distance (3ft). Microphone is Minidsp UMIK-1 So, these are nearfield measurements, tomorrow I try to run farfield/room measurements.

Both programs use a sine weep as signal and record it's impulse response for analysis. REW uses eq-file for the mic, TDA doesn't.

SPL/Frequency some differencies. I don't know what IR gating and smoothing TDA uses, REW is set for 500ms ERB here.

TDA vs. wavelet transform presentation of delayed energy. Notice different nominators for axes, and different time scale.
Attached Images
File Type: jpg dtm 1m spl-vert.jpg (122.2 KB, 408 views)
File Type: jpg dtm TDA wavelet -vert.jpg (129.1 KB, 405 views)
__________________
AINOgradient speaker project
  Reply With Quote
Old 20th January 2016, 09:04 PM   #3
wesayso is offline wesayso  Netherlands
diyAudio Member
 
wesayso's Avatar
 
Join Date: Jul 2007
As I brought this subject up I'll join the discussion and post my own plots here. I didn't save the DFR (demo required screen grabs)
The interest in this program was generated by Mr. Raimonds Skuruls post in the full range forum: The room correction or speaker correction? What can we do with dsp power now availabl

As time coherency was high on my agenda searching a way to correct my speaker response I decided to run this software in demo mode.

First I'll show the REW equivalent:
Click the image to open in full size.
Wavelet view in REW of left + right speaker at listening position.

On to the APL_TDA measurement at that same position:
Click the image to open in full size.
3D wavelet view of the APL_TDA software

Click the image to open in full size.
2D view of that same measurement. Figuring out how my DFR plot looked based on this view wouldn't be to hard.

If I run my arrays without any correction (beside some very global PEQ values) I get the following result:
Click the image to open in full size.
Same position, again left plus right speaker, this time without my FIR based correction.

As my speakers are full range, floor to ceiling line arrays the plan from the start was to correct these with FIR based EQ. I've spend quite a bit of time to find a recipe that worked for me, concentrating on the least amount of correction with the best result I could get.
To get there I treated first reflection points with damping panels. From there I used REW's different views on the impulse (all of the tabs available) to see what could be done theoretically. I watched what happens over time by using different gating lengths in all those tabs and measured at more than one point to see big trends in the frequency response at different places in the room. Based on that information I setup DRC with a custom template and values to correct what I wanted it to correct using the left and right speaker measurement at the sweet spot. There are still some room effects in the separate left or right measurements. As can be seen in this plot:
Click the image to open in full size.
The results of the left channel, right channel and it's combined response as measured at the listening position. This plot is gated with a 4 cycle frequency dependant window in REW. It is possible to run DSP based on longer frequency dependant windows to eliminate all those remaining deviations. But the key to success was to figure out how to separate the global results happening because of the specific speaker type and those caused by the room. As those room effects change too much with position to correct them with DSP. This is my attempt to correct the direct speaker response, without trying to correct what happens in the room later in time.

As said, I got my results based on my research of REW measurements and final correction with DRC-FIR. JRiver is my convolution engine and is used to apply additional PEQ edits. The combined result can be seen in these plots.
I merely tested the demo version of APL_TDA out of curiosity.

Last edited by wesayso; 20th January 2016 at 09:12 PM.
  Reply With Quote
Old 21st January 2016, 09:00 AM   #4
Juhazi is offline Juhazi  Finland
diyAudio Member
 
Juhazi's Avatar
 
Join Date: Mar 2007
Location: Jyväskylä, Finland
Okay, I two hours of my precious free time to get comparable measurements with REW and TDA.
Output: Laptop-usb-odac-analog-minidsp-amplifiers-speaker drivers(4-way AINOgradient)
Input: UMIK-1-usb-laptop
Measuring distance roughly 9ft (2,7m) LEFT speaker at normal position 2ft from wall (which makes the rising bass response) Microphone on stand at 3ft height 4ft from backwall. Sidewalls 10-15ft away.

SPL response - Similar, if REW is set to 500ms and ERB smoothing
Wavelet vs. TDA3D analysis - Similar except for graphic presentation details
DFR TDA can show delays in a simple way - the room makes the humps
Attached Images
File Type: jpg ainogneo83 v7b L tda3d.jpg (125.8 KB, 54 views)
File Type: jpg ainogneo83 v7b L tda3d 200ms.jpg (122.7 KB, 53 views)
File Type: jpg ainogneo83 v7b L afr.jpg (127.9 KB, 68 views)
File Type: jpg ainogneo83 v7b L dfr.jpg (122.6 KB, 65 views)
__________________
AINOgradient speaker project

Last edited by Juhazi; 21st January 2016 at 09:03 AM.
  Reply With Quote
Old 21st January 2016, 09:02 AM   #5
Juhazi is offline Juhazi  Finland
diyAudio Member
 
Juhazi's Avatar
 
Join Date: Mar 2007
Location: Jyväskylä, Finland
Measurements with REW
Attached Images
File Type: jpg ainogneo83 v7b lr 7ms nosmo.jpg (66.4 KB, 66 views)
File Type: jpg ainogneo83 v7b LR 500ms erb.jpg (53.3 KB, 48 views)
File Type: jpg ainogneo83 v7b L wavelet default.jpg (76.8 KB, 58 views)
File Type: jpg ainogneo83 v7b L decay.jpg (113.9 KB, 64 views)
__________________
AINOgradient speaker project
  Reply With Quote
Old 21st January 2016, 09:12 AM   #6
diyAudio Member
 
eriksquires's Avatar
 
Join Date: May 2013
Very cool stuff. A very small part of it reminds me of the things Meyer's Sound Labs was doing a few years ago, but they were using live music to achieve only some of the results here.

Advantages that approach had was the ability to continually tune the auditorium as people arrived. I really wish I had not blown my budget on the Organic Wax and Saffron capacitors already.

Best,


Erik
__________________
All of my successes and failures can be explained by my attempts at getting a girlfriend. All of them.
  Reply With Quote
Old 21st January 2016, 10:49 AM   #7
wesayso is offline wesayso  Netherlands
diyAudio Member
 
wesayso's Avatar
 
Join Date: Jul 2007
I should probably post my 7ms unsmoothed window:
Click the image to open in full size.

And a 500 ms ERB plot as well:
Click the image to open in full size.

And the Spectrogram of left the left side:
Click the image to open in full size.

Quite daring to plot it out to -40 dB in a living room but in the interest of this study I'll comply.

Last edited by wesayso; 21st January 2016 at 11:08 AM.
  Reply With Quote
Old 21st January 2016, 11:21 AM   #8
Juhazi is offline Juhazi  Finland
diyAudio Member
 
Juhazi's Avatar
 
Join Date: Mar 2007
Location: Jyväskylä, Finland
Wesayso's Line Arrays shine at low to mid F up to 3kHz, and not bad at all even higher. FIR based eq is applied and it works specially in lower octaves against room modes.
__________________
AINOgradient speaker project
  Reply With Quote
Old 21st January 2016, 11:26 AM   #9
wesayso is offline wesayso  Netherlands
diyAudio Member
 
wesayso's Avatar
 
Join Date: Jul 2007
My left side is having a corner problem as is evident.
The right side is doing better there but has it's own problems with reflections around the 100-200 Hz. But without turning my living room into a studio the FIR filtering works very well to get things in check.

Fix all room problems with FIR and it starts to sound funny (possibly giving FIR based processing a bad name). The key here is to apply no more correction than needed. That sounds easier than it is though. Took me almost a year to find out.
  Reply With Quote
Old 21st January 2016, 02:34 PM   #10
BYRTT is offline BYRTT  Denmark
diyAudio Member
 
Join Date: Dec 2009
Not sure i dare

Click the image to open in full size.

Thanks cool thread and data : )
Attached Images
File Type: png 5.PNG (258.0 KB, 372 views)
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Software for Acoustic Simulation masviper Software Tools 0 7th September 2011 07:02 PM
Anyone know of Alex Peychev of APL HiFi and his new ultimate CDP? MGH Digital Source 5 25th May 2009 10:39 PM
Acoustic Modeling Software HK26147 Multi-Way 0 24th November 2008 12:53 PM
HELP!! 3D Acoustic design software needed Avetik Multi-Way 3 26th December 2006 07:05 PM
L.a.d.p. 2 Loudspeaker Software.... ermes Multi-Way 2 6th July 2004 03:46 PM


New To Site? Need Help?

All times are GMT. The time now is 11:07 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright ©1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2
Wiki